About the Intricacy of Tasks

Leonard M. Eberding; Matteo Belenchia; Arash Sheikhlar; Kristinn R. Thórisson

doi:10.1007/978-3-030-93758-4_8

About the Intricacy of Tasks

Leonard M. Eberding^*, Matteo Belenchia, Arash Sheikhlar, Kristinn R. Thórisson

^*Corresponding author for this work

Department of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Without a concrete measure of the “complicatedness” of tasks that artificial agents can reliably perform, assessing progress in AI is difficult. Only by providing evidence of progress towards more complicated tasks can developers aiming for general machine intelligence (GMI) ascertain their progress towards that goal. No such measure for this exists at present. In this work we propose a new measure of the intricacy of tasks, especially designed to describe their physical composition and makeup. Our intricacy is a multi-dimensional measurement that depends purely on objective physical properties of tasks and the environment in which they are to be performed. From this task intricacy measure, a relation to the knowledge of learners can allow calculation of the difficulty of a particular task for a particular learner. The method is intended for both narrow-AI and GMI-aspiring systems. Here we discuss some of the implications of our intricacy measure and suggest ways in which it may be used in AI research and system evaluation.

Original language	English
Title of host publication	Artificial General Intelligence - 14th International Conference, AGI 2021, Proceedings
Editors	Ben Goertzel, Matthew Iklé, Alexey Potapov
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	65-74
Number of pages	10
ISBN (Print)	9783030937577
DOIs	https://doi.org/10.1007/978-3-030-93758-4_8
Publication status	Published - 2022
Event	14th International Conference on Artificial General Intelligence, AGI 2021 - San Francisco, United States Duration: 15 Oct 2021 → 18 Oct 2021

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	13154 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	14th International Conference on Artificial General Intelligence, AGI 2021
Country/Territory	United States
City	San Francisco
Period	15/10/21 → 18/10/21

Bibliographical note

Funding Information:
Acknowledgments. This work was supported in part by Cisco Systems, the Icelandic Institute for Intelligent Machines and Reykjavik University.

Publisher Copyright:
© 2022, Springer Nature Switzerland AG.

Other keywords

Artificial intelligence
Difficulty
Environments
Evaluation
General machine intelligence
Intricacy
Task theory
Tasks
Training

Access to Document

10.1007/978-3-030-93758-4_8

Cite this

Eberding, L. M., Belenchia, M., Sheikhlar, A., & Thórisson, K. R. (2022). About the Intricacy of Tasks. In B. Goertzel, M. Iklé, & A. Potapov (Eds.), Artificial General Intelligence - 14th International Conference, AGI 2021, Proceedings (pp. 65-74). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13154 LNAI). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-93758-4_8

Eberding, Leonard M. ; Belenchia, Matteo ; Sheikhlar, Arash et al. / About the Intricacy of Tasks. Artificial General Intelligence - 14th International Conference, AGI 2021, Proceedings. editor / Ben Goertzel ; Matthew Iklé ; Alexey Potapov. Springer Science and Business Media Deutschland GmbH, 2022. pp. 65-74 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{21273e9b0429440f8206834a83434513,

title = "About the Intricacy of Tasks",

abstract = "Without a concrete measure of the “complicatedness” of tasks that artificial agents can reliably perform, assessing progress in AI is difficult. Only by providing evidence of progress towards more complicated tasks can developers aiming for general machine intelligence (GMI) ascertain their progress towards that goal. No such measure for this exists at present. In this work we propose a new measure of the intricacy of tasks, especially designed to describe their physical composition and makeup. Our intricacy is a multi-dimensional measurement that depends purely on objective physical properties of tasks and the environment in which they are to be performed. From this task intricacy measure, a relation to the knowledge of learners can allow calculation of the difficulty of a particular task for a particular learner. The method is intended for both narrow-AI and GMI-aspiring systems. Here we discuss some of the implications of our intricacy measure and suggest ways in which it may be used in AI research and system evaluation.",

keywords = "Artificial intelligence, Difficulty, Environments, Evaluation, General machine intelligence, Intricacy, Task theory, Tasks, Training",

author = "Eberding, {Leonard M.} and Matteo Belenchia and Arash Sheikhlar and Th{\'o}risson, {Kristinn R.}",

note = "Funding Information: Acknowledgments. This work was supported in part by Cisco Systems, the Icelandic Institute for Intelligent Machines and Reykjavik University. Publisher Copyright: {\textcopyright} 2022, Springer Nature Switzerland AG.; 14th International Conference on Artificial General Intelligence, AGI 2021 ; Conference date: 15-10-2021 Through 18-10-2021",

year = "2022",

doi = "10.1007/978-3-030-93758-4_8",

language = "English",

isbn = "9783030937577",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "65--74",

editor = "Ben Goertzel and Matthew Ikl{\'e} and Alexey Potapov",

booktitle = "Artificial General Intelligence - 14th International Conference, AGI 2021, Proceedings",

address = "Germany",

}

Eberding, LM, Belenchia, M, Sheikhlar, A & Thórisson, KR 2022, About the Intricacy of Tasks. in B Goertzel, M Iklé & A Potapov (eds), Artificial General Intelligence - 14th International Conference, AGI 2021, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13154 LNAI, Springer Science and Business Media Deutschland GmbH, pp. 65-74, 14th International Conference on Artificial General Intelligence, AGI 2021, San Francisco, United States, 15/10/21. https://doi.org/10.1007/978-3-030-93758-4_8

About the Intricacy of Tasks. / Eberding, Leonard M.; Belenchia, Matteo; Sheikhlar, Arash et al.
Artificial General Intelligence - 14th International Conference, AGI 2021, Proceedings. ed. / Ben Goertzel; Matthew Iklé; Alexey Potapov. Springer Science and Business Media Deutschland GmbH, 2022. p. 65-74 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13154 LNAI).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - About the Intricacy of Tasks

AU - Eberding, Leonard M.

AU - Belenchia, Matteo

AU - Sheikhlar, Arash

AU - Thórisson, Kristinn R.

N1 - Funding Information: Acknowledgments. This work was supported in part by Cisco Systems, the Icelandic Institute for Intelligent Machines and Reykjavik University. Publisher Copyright: © 2022, Springer Nature Switzerland AG.

PY - 2022

Y1 - 2022

N2 - Without a concrete measure of the “complicatedness” of tasks that artificial agents can reliably perform, assessing progress in AI is difficult. Only by providing evidence of progress towards more complicated tasks can developers aiming for general machine intelligence (GMI) ascertain their progress towards that goal. No such measure for this exists at present. In this work we propose a new measure of the intricacy of tasks, especially designed to describe their physical composition and makeup. Our intricacy is a multi-dimensional measurement that depends purely on objective physical properties of tasks and the environment in which they are to be performed. From this task intricacy measure, a relation to the knowledge of learners can allow calculation of the difficulty of a particular task for a particular learner. The method is intended for both narrow-AI and GMI-aspiring systems. Here we discuss some of the implications of our intricacy measure and suggest ways in which it may be used in AI research and system evaluation.

AB - Without a concrete measure of the “complicatedness” of tasks that artificial agents can reliably perform, assessing progress in AI is difficult. Only by providing evidence of progress towards more complicated tasks can developers aiming for general machine intelligence (GMI) ascertain their progress towards that goal. No such measure for this exists at present. In this work we propose a new measure of the intricacy of tasks, especially designed to describe their physical composition and makeup. Our intricacy is a multi-dimensional measurement that depends purely on objective physical properties of tasks and the environment in which they are to be performed. From this task intricacy measure, a relation to the knowledge of learners can allow calculation of the difficulty of a particular task for a particular learner. The method is intended for both narrow-AI and GMI-aspiring systems. Here we discuss some of the implications of our intricacy measure and suggest ways in which it may be used in AI research and system evaluation.

KW - Artificial intelligence

KW - Difficulty

KW - Environments

KW - Evaluation

KW - General machine intelligence

KW - Intricacy

KW - Task theory

KW - Tasks

KW - Training

UR - http://www.scopus.com/inward/record.url?scp=85123302602&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-93758-4_8

DO - 10.1007/978-3-030-93758-4_8

M3 - Conference contribution

AN - SCOPUS:85123302602

SN - 9783030937577

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 65

EP - 74

BT - Artificial General Intelligence - 14th International Conference, AGI 2021, Proceedings

A2 - Goertzel, Ben

A2 - Iklé, Matthew

A2 - Potapov, Alexey

PB - Springer Science and Business Media Deutschland GmbH

T2 - 14th International Conference on Artificial General Intelligence, AGI 2021

Y2 - 15 October 2021 through 18 October 2021

ER -

Eberding LM, Belenchia M, Sheikhlar A, Thórisson KR. About the Intricacy of Tasks. In Goertzel B, Iklé M, Potapov A, editors, Artificial General Intelligence - 14th International Conference, AGI 2021, Proceedings. Springer Science and Business Media Deutschland GmbH. 2022. p. 65-74. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-93758-4_8

About the Intricacy of Tasks

Abstract

Publication series

Conference

Bibliographical note

Other keywords

Access to Document

Fingerprint

Cite this